[Wien] Electric Field -- is energy right?

[Gavin Abo]

Not sure if it helps, but below are notes I have made on the electric field:
--------------------------------------------------------------------------------------------
Use of the electric field seems somewhat undocumented.

Reference the case.in0 section (7.1.3) in the WIEN2k usersguide [ 
http://susi.theochem.tuwien.ac.at/reg_user/textbooks/usersguide.pdf ].

There seems to be 13 modes and 2 Fourier options to choose from. To list 
the modes, add:

-999 2

on line 4 in the case.in0 file, then run in a terminal in the case 
directory:

x lapw0

It should output:

Available E-field modes
Mode  0 is Analytic triangular ramp
Mode  1 is =0 if |z|<lambda, otherwise E*cos((z0-lambda)*pi2/(0.5-lambda))
Mode  2 is =0 if |Z|<lambda, otherwise 
E*(1.D0-cos((z0-lambda)*pi2/(0.5-lambda)))
Mode  3 is =0 if |z|<lambda, otherwise 
E*(1-cos((z0-lambda)*pi/(0.5-lambda)))*0.5D0
Mode  4 is =E if |z|<lambda, otherwise E*(0.5D0-Z0)/(0.5d0-lambda)
Mode  5 is =0 if |z|<lambda, otherwise E*(1.D0-(0.5D0-Z0)/(0.5d0-lambda))
Mode  6 is Form from Lozovoi et al, J. Chem Phys, 115, 1661, 2001, 
normalized to max of 1
Mode  7 is Form from Lozovoi et al, J. Chem Phys, 115, 1661, 2001
Mode  8 is =Constant E
Mode  9 is =0 if |z|<lambda, otherwise E
Mode 10 is =E if |z|<lambda, otherwise 
E*((1-cos((z0-lambda)*pit/cut))*0.5D0)
Mode 11 is Numerical Triangular ramp from E/2 to -E/2
Mode 12 is Numerical Triangular ramp from 0 to -E
Add 20 for finite Fourier series
Add 40 for Fourier expansion within RMTs

Note: The above list isn't fully displayed when x lapw0 is ran in w2web, 
so use a terminal instead of w2web.

According to the post

http://zeus.theochem.tuwien.ac.at/pipermail/wien/2008-May/010928.html

run a typically scf cycle to convergence without the electric field 
(line 4 blank in case.in0).

Then, add the electric field.  For example, if "Mode 8 is =Constant E" 
is chosen with an amplitude of 5 Ry, then line 4 of case.in0 would be

8000 5

where Mode = IEFIELD/1000 = 8000/1000 = 8

If a value of lambda of say 3 is specified such as for Mode 1, then add 
it as a third column value to line 4 of case.in0:

1000 5 3

Then, run x lapw0 or another scf cycle for the electric field.

I don't know if a supercell needs to be used for all or just some of the 
potentials.  Could try asking the mailing list.  Also, refer to the posts:

http://zeus.theochem.tuwien.ac.at/pipermail/wien/2011-June/014946.html
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2004-September/003549.html
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2008-September/011449.html
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2008-September/011340.html

E-field source code: SRC_lapw0/epot1.f
--------------------------------------------------------------------------------------------

As a quick test only, I put "120 -0.50" in line 4 of TiC.in0 and ran x 
lapw0.  The output in TiC.scf0 suggests the 120 gives a triangular ramp 
electric field with 120 Fourier coefficients, and it also sets an 
amplitude of -0.5 Ry.  So I'm wondering if the behavior seen in the 
difference observed between structures A & B has anything to due with 
how the triangular ramps have been positioned in those two structures.

username at computername:~/wiendata/TiC$ grep :EFIELD *.scf0
:EFIELD       E-field of   -0.50000 Ry with  120 Fourier coefficients 
requested
:EFIELD       Form is Analytic triangular ramp
:EFIELD       Number of FC in E-field   8 , Largest   14

On 3/7/2017 8:59 AM, Laurence Marks wrote:
> I probably should know/remember the answer to this question, but I don't.
>
> In lapw0 we can add an electric field, is the energy right or not?
>
> For reference, I have two different structures A & B and I want to 
> compare the change in energy between the two in a field. The 
> difference is opposite to what I expected, so either my expectation is 
> wrong or an energy term is missing. For reference, I have added "120 
> -0.50" at the bottom of case.in0 and I am minimizing the 
> positions/forces (which I seem to remember were right but that does 
> not prove that a constant is not missing). Uncharged unit cell.
>
>
> -- 
> Professor Laurence Marks
> "Research is to see what everybody else has seen, and to think what 
> nobody else has thought", Albert Szent-Gyorgi
> www.numis.northwestern.edu <http://www.numis.northwestern.edu> ; 
> Corrosion in 4D: MURI4D.numis.northwestern.edu 
> <http://MURI4D.numis.northwestern.edu>
> Partner of the CFW 100% program for gender equity, 
> www.cfw.org/100-percent <http://www.cfw.org/100-percent>
> Co-Editor, Acta Cryst A
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20170307/2e06d399/attachment.html>